US7779215B2 - Method and related apparatus for accessing memory - Google Patents

Method and related apparatus for accessing memory Download PDF

Info

Publication number
US7779215B2
US7779215B2 US10/906,748 US90674805A US7779215B2 US 7779215 B2 US7779215 B2 US 7779215B2 US 90674805 A US90674805 A US 90674805A US 7779215 B2 US7779215 B2 US 7779215B2
Authority
US
United States
Prior art keywords
memory
rank
channel
data
accessing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
US10/906,748
Other versions
US20050289317A1 (en
Inventor
Ming-Shi Liou
Bowei Hsieh
Jiin Lai
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Via Technologies Inc
Original Assignee
Via Technologies Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Via Technologies Inc filed Critical Via Technologies Inc
Assigned to VIA TECHNOLOGIES INC. reassignment VIA TECHNOLOGIES INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: HSIEH, BOWEI, LAI, JIIN, LIOU, MING-SHI
Publication of US20050289317A1 publication Critical patent/US20050289317A1/en
Application granted granted Critical
Publication of US7779215B2 publication Critical patent/US7779215B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/16Handling requests for interconnection or transfer for access to memory bus
    • G06F13/1668Details of memory controller
    • G06F13/1684Details of memory controller using multiple buses
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C2207/00Indexing scheme relating to arrangements for writing information into, or reading information out from, a digital store
    • G11C2207/22Control and timing of internal memory operations
    • G11C2207/2281Timing of a read operation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Dram (AREA)
  • Transceivers (AREA)

Abstract

A method for utilizing the multi-channel transmission bandwidth in an asymmetrically arranged memory is provides. The present invention defines symmetrically arranged parts of the memory ranks of the memory as a virtual ranks. If data is stored in symmetrically arranged memory ranks of the memory, channels corresponding to the symmetrically arranged memory ranks could be simultaneously utilized to transfer data. If data is stored in an asymmetrically arranged memory rank of the memory, the channel corresponding to the asymmetrically arranged memory rank could only be utilized to transfer data.

Description

BACKGROUND OF INVENTION
1. Field of the Invention
The invention relates to a memory accessing method and related memory accessing device, and more particularly, to a memory accessing method and related memory accessing device capable of increasing accessing efficiency in a asymmetric multi-channel memory device.
2. Description of the Prior Art
Microprocessor systems, such as computer systems, have become one of the most important hardware devices in today's modern information society. Therefore, demands for increasing the efficiency of the microprocessor system have become greater and greater. As known by those of ordinary skill in the art, the computer system is operated through the cooperation of a central processing unit (CPU), a chip set (such as implemented by north-bridge chips), and a main memory. For example, the main memory can be a random accessing device for storing program codes and data. The CPU accesses and executes the program codes and the related data through the chip set to achieve functions of the computer system. As mentioned above, it is well-known that the key point to increase the efficiency of the computer/microprocessor system is to increase the accessing efficiency of the main memory.
It is also known by those skilled in the art that the main memory includes one or several memory modules in a typical computer system. Each memory module could be with one or two memory ranks, and each memory rank could be with a plurality of memory devices to store data. For example, in the present computer system standard, a single-sided or double-sided dual-inline memory module (DIMM) is disclosed. The single-sided memory module has memory devices on one side, and these memory devices are connected as a memory rank. Conversely, the double-sided memory module has memory devices on both two sides, and memory devices of each side are respectively connected to a memory rank. Therefore, the double-sided memory module can provide two memory ranks with the same memory capacity. The present computer system could include a plurality of memory modules. These memory modules could be integrated as a single memory through the chipset control so that the memory modules are integrated to a memory capacity that supports the computer system.
SUMMARY OF INVENTION
The invention provides a memory accessing method and related memory accessing apparatus with capable of utilizing the characteristic of a double-channel in an asymmetrically arranged memory, to solve the above-mentioned problem.
A multi-channel memory accessing method of the present invention, each channel respectively connects to at least one memory rank, and the accessing method including: determining whether a target memory rank of accessing data belongs to a virtual rank; if the target memory rank does not belong to the virtual rank, utilizing a predetermined transmission bandwidth to access the data; and if the target memory rank belongs to the virtual rank, utilizing times of the predetermined transmission bandwidth to access the data.
A multi-channel memory accessing apparatus of the present invention including: a controlling chip; a plurality of memory modules, wherein each memory module comprises at least one memory rank; and a plurality of channels, wherein each channel respectively corresponds to one of the memory modules for connecting the memory module and the controlling chip and transferring data of the at least one memory rank corresponding to the memory module; wherein if there respectively has a first memory rank with same memory capacity on each channels, the controlling chip defines the plurality of first memory ranks on each channels as a virtual rank.
These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a block diagram of a symmetrically positioned memory device in a computer system according to the present invention.
FIG. 2 is a diagram of the memory device shown in FIG. 1 positioned symmetrically.
FIG. 3 is a timing chart of each related data signal when the chip set accesses data from the memory device according to the symmetrically positioned memory device shown in FIG. 2.
FIG. 4 is a diagram of a non-symmetrically positioned memory device in a computer system according to the present invention.
FIG. 5 is a timing chart of each related signal when the non-symmetric memory device shown in FIG. 4 is accessed according to the present invention.
FIG. 6 and FIG. 7 are diagrams of two other symmetrically positioned memory devices according to the present invention.
FIG. 8 is a flow chart of accessing data from a memory according to the present invention.
DETAILED DESCRIPTION
Please refer to FIG. 1, which is a block diagram of a controller 20 applied in a chipset 38 of a computer system 42 to perform accessing control of a double-channel memory 30 according to the present invention. In the computer system 42 (for example, a personal computer or other microprocessor systems), the chipset 38 could be a memory accessing control chip, being as an accessing interface of the memory 30. Therefore, a host 40 could access memory 30 through the above-mentioned chipset 38. The host 40 could be a central processing unit (CPU), other devices or circuits with capable of performing direct memory accessing (DMA) operations.
Furthermore, the chipset 38 includes an interface module 22 and a controller 20. In addition, in order to achieve a technique of the present invention, the controller 20 has a determining module 24, an accessing module 26, and an addressing module 28. The memory 30 has a plurality of slots (as 34A, 34B, 36A, and 36B shown in FIG. 1), a memory module (for example, a random accessing double-sided memory module, DIMM) could be plugged in one slot. Memory capacity of memory 30 providing to the host 40 depends on total memory modules plugged into the slots. The chipset 38, the channel 32A, 32B and each memory module inside the memory 30 are regarded as a multi-channel (e.g., a double-channel) memory accessing apparatus.
In order to access the data stored in the memory module plugged in each slot, the accessing module 26 transfers instructions and signals to access the memory modules. If a double-sided memory module with two memory ranks is plugged in one slot, the slot could respectively transfer control signals to corresponding different memory ranks on the same memory module. In addition, each memory module could transfer the data stored in the memory module through the channels connected to the certain slot to respond the access control instructions and signals.
As shown in FIG. 1, the slot 34A and 36A are connected to the channel 32A, and the slot 34B and 36B are connected to the channel 32B. Therefore, the memory module plugged in the slot 34A or 36A could transfer data through the channel 32A. Similarly, the memory module plugged in the slot 34B or 36B could transfer data through the channel 23B.
A more detailed description of how the chip set 38 performs the access controlling operation on the memory device 30 is illustrated as follows.
When the computer system 42 is turned on, the computer system 42 tests the memory 30 (for example, the computer system 42 executes the basic input output system (BIOS) to test the memory 30) so that the computer system 42 could detect the position/configuration of each memory module inside the memory 30. For example, the memory capacity of each memory module plugged in each slot could be detected by the computer system 42 through the above-mentioned operation. Therefore, the computer system 42 ensures the whole memory capacity is provided by the memory 30. The chipset 38 addresses the whole memory capacity provided by the memory 30. That is, the chipset 38 defines the memory 30 as different addresses (such as the physical addresses) and establishes a mapping relationship between the addresses and the memory ranks. Thus the host 40 could randomly access data of the memory 30. For example, assuming that memory modules installed in the memory 30 could totally provide three memory ranks, the chipset 38 maps memory space, provided by the three memory ranks, to different addresses. When the host 40 has to store data into a given address, the chipset 38 determines the target memory rank corresponding to the given address, then the chipset 38 utilizes control signals to transfer data into the corresponding memory rank and thereby store data. Similarly, if the host 40 has to read data from a certain address, the chipset 38 determines the target memory rank corresponding to the given address, and then the chipset 38 utilize the corresponding control signals to read data from the memory rank.
The present invention could also perform different modes to access memory 30 according to the arrangement of memory 30.
The first embodiment of the present invention is shown in FIG. 2 and FIG. 2. FIG. 2 is a diagram illustrating a memory access apparatus 42, wherein the memory 30 is symmetrically arranged. In this embodiment shown in FIG. 2, assuming that double- sided memory modules 46A and 46B are respectively plugged in the slots 34A and 34B. The memory module 46A has two memory ranks labeled 460 and 461, and the memory module 46B also has two memory ranks labeled 464 and 465. The memory ranks 460 and 461 receive the corresponding control signals S0 and S1 respectively through the slots 34A; and the memory ranks 464 and 465 receive the corresponding control signals S4 and S5 respectively through the slots 34B.
In FIG. 2, Assuming the memory rank 460, which connects to the channel 32A, has same memory capacity with compared to the memory rank 464, which connects to the channel 32B; also, assuming the memory rank 461, which connects to the channel 32A, has same memory capacity with compared to the memory rank 465, which connects to the channel 32B. As a result, the memory 30 is regarded as a double-channel symmetrically arranged memory. When the memory 30 is symmetrically arranged, controller 20 of the present invention could define the symmetrically arranged memory ranks on the double-channel as a virtual memory rank, and access the memory 30 through the double-channel transmission bandwidth.
As FIG. 2 shows, due to symmetrically arranged memory, the addressing module 28 would define the memory ranks 460 and 464 as a virtual rank, and defines the memory ranks 461 and 465 as another virtual rank. The whole memory capacity provided by the memory 30 could be mapped to corresponding addresses.
When storing data (for example, 8 QWs of data, where one QW (quad-word) is 16 bits) into memory 30, then the accessing module 26 would divide the data into two parts (e.g., each part having 4 QW). These two parts of data are respectively stored into the symmetric memory ranks, which are defined as same virtual rank, through the two channels. For example, one part of data is stored into the memory rank 460 (or 461), another part of data is stored into memory rank 464 (or 465). Because the two parts of data are respectively transferred to corresponding memory ranks at the same time through the two channels, as a result, the double-channel transmission bandwidth is utilized the same time.
Similarly, when read the data from the target virtual rank, the accessing module 26 would respectively read the two parts of data from symmetric memory ranks (ex. from memory ranks 460 and 464 or from memory ranks 461 and 465) through the two channels and interleave the two parts of data into the original data.
FIG. 3 is a timing chart of each related data signal when the chipset 38 accesses data from the memory 30 according to the symmetrically arranged memory 30 shown in FIG. 2. And please note that the horizontal axis shown in FIG. 3 shows time.
Assuming that the host 40 requests to read a 8 QW data, which is composed of eight data blocks Dt0 to Dt7 and each one data block stands for 1 QW, of a certain address from the memory 30 at time t0. Assuming the mapping address is located in the virtual rank formed by the memory ranks 460 and 464. The accessing module 26 simultaneously utilizes similar reading controlling instructions RDX as control signals S0 and S4 (shown in FIG. 2) at time t1, and then accessing module 26 respectively transfers the control signals S0 and S4 to the memory ranks 460 and 464 with connected to the two channels. At time t2, the memory rank 460 orderly transfers the data block Dt0, Dt2, Dt4, and Dt6 through the channel 32A. Simultaneously the memory rank 464 orderly transfers the data blocks Dt1, Dt3, Dt5, and Dt7 through the channel 32B. The chipset 38 interleaves data blocks Dt0 to Dt7 into the original 8 QW data. In other words, when an 8 QW data has to be stored into the symmetrically arranged memory 30, the chipset 38 could symmetrically control the symmetric memory ranks of the two channels and respectively stores 4 QW data in each memory rank. This is, the chipset 38 could divide the 1st, 3rd, 5th, and 7th QW as one part of the 8 QWs data and divide the 2nd, 4th, 6th, and 8th QW as another part of the 8 QWs data then respectively stores the divided two parts of 8 QWs data into two memory ranks, which is regarded as a virtual rank. If the 8 QWs data has to be read from the virtual rank of the symmetrically arranged memory device 30, the chipset 38 could respectively read data blocks from the two memory ranks (each memory rank has four data blocks in it) of the virtual rank through the two channels; and interleaves these eight data blocks into the original 8 QW data.
As mentioned above, if the memory 30 is symmetrically arranged memory, the chipset 38 could utilize the same control instruction and control signal to respectively control the symmetric memory ranks of the double-channel to access data. Thus the 8 QWs data could be transferred on the two channels (double-channel) at the same time. Comparing to the prior art, it only takes two period times (2T) to complete the 8 QWs data transmission (assuming that the memory module is a double-data-rate memory module). Furthermore, the double-channel symmetric accessing could reduce the accessing latency. For example, at time t3, the accessing module 26 could transfer another control instruction RDY as the control signal so that the symmetric memory ranks can utilize the double-channel to transfer another 8 QWs data Dy0 to Dy7 in order to transfer another 8 QWs data.
But if the memory is asymmetrically arranged, the foregoing mention method with high transmission bandwidth could not be utilized. If the memory is asymmetrically arranged, the chipset could not utilize symmetric control signals to control different memory ranks on the different channels at the same time. Regardless of which memory rank the data is stored in, the chipset could only access one memory rank at a time and transfer data through one channel at a time. In other words, one 8 QWs data can only be transferred through one channel. That is, four period times (4T) are necessary to complete the 8 QWs data transmission.
The present invention provides a method and an apparatus for utilizing the multi-channel transmission bandwidth in an asymmetrically arranged memory, so that data transmission efficiency could be increased. In the present invention, even when a memory is asymmetrically arranged, parts of the symmetrically arranged memory ranks are still defined as a virtual rank, as a result, the multi-channel transmission bandwidth could be applied. Taking double-channel memory as examples, more detail description is illustrated as follows.
The second embodiment of the present invention is shown in FIG. 4 and FIG. 5. Assuming that the memory 30 in FIG. 4 is an asymmetrically arranged memory. FIG. 5 is a timing chart of related data signals of accessing the memory shown in FIG. 4 according to the present invention. The horizontal axis of FIG. 5 is time axis.
Assuming the asymmetrically arranged memory 30, shown in FIG. 4, has three slots 34A, 36A and 34B with corresponding memory modules 46A, 48A and 46B plugged in respectively. The memory module 46A and 46B both are double-sided memory module. The memory module 46A has memory ranks 460 and 461 and the memory module 48A has memory ranks 464 and 465.Assuming the memory ranks 460 and 464 in two channels 32A and 32B with the same memory capacity and the memory ranks 461 and 465 in two channel 32A and 32B with the same memory capacity are regarded as symmetric memory ranks and are the symmetrical parts of the asymmetrically arranged memory 30. Alternatively, the double-sided memory module 48A having memory ranks 462 and 463 in channel 32A is the asymmetrical parts of the asymmetrically arranged memory 30, due to the memory ranks 462 and 463 have no corresponding memory ranks in another channel 32B.
Although the memory 30 is asymmetric, the memory ranks 460 and 461 have corresponding memory ranks 464 and 465 with the same memory capacity in another channel; as a result, the addressing module 28 could define the memory ranks 460 and 464 as a virtual rank. The virtual ranks could be accessed in the double-channel data transmission bandwidth. The remained memory ranks 462 and 463 in channel 32A have no corresponding memory ranks with same capacity in channel 32B, so memory ranks 462 and 463 could only be accessed in one-channel data transmission bandwidth.
In the present invention when the computer system 42 is turned on and detects that the memory 30 is asymmetrically arranged, addressing module 28 defines symmetrically arranged parts of memory ranks of the memory 30 on the double-channel as a virtual rank. Furthermore, the addressing module 28 sets a mapping relationship according to the above-mentioned arranged relationship of the memory ranks. When the host 40 writes data into a certain address of the memory 30, the determining module 24 first determines which memory space (the virtual ranks of the double-channel or the memory ranks of the single channel) that the address is mapped to. If the address is mapped to the memory space of the virtual ranks, the accessing module 26 could utilize the same control signal to control the symmetric memory ranks and utilize the double-channel to the symmetric memory ranks to transfer different parts of the data so that the data could be transferred in the double-channel data transmission bandwidth. Alternatively, if the address is mapped to the memory space of asymmetrically arranged memory ranks, the accessing module 26 could transfer the data to the same memory rank in one-channel transmission bandwidth.
Please refer to FIG. 4 and FIG. 5. Assuming that the host 40 writes two 8 QW data to memory 30, and each 8 QW data is composed of eight data blocks. If the address of the first 8 QW data (Dt0-Dt7) is mapped to the virtual rank of the symmetrically arranged memory ranks 461 and 465 shown in FIG. 4, thus the first 8 QW data (Dt0-Dt7) is divided into two parts and respectively stored in the memory ranks 461 and 465. If the address of the second 8 QW data (Dy0-Dy7) is mapped to the memory ranks 462, which has no symmetrically arranged memory rank on another channel, thus the second 8 QW data (Dy0-Dy7) is stored in the memory rank 462.
When the host 40 has to access the above-mentioned two pieces of data, the determining module 24 determines the addresses to which the two data are mapped to. When accessing the first 8 QW data, the determining module 24 determines that the first 8 QW data is stored in the virtual rank of the symmetrically arranged memory ranks 461 and 465. Then the accessing module 26 simultaneously utilizes the same control instruction RDX as the control signals S1 and S5 of channel 32A and 32B to sequentially transfer data blocks Dt0, Dt2, Dt4, Dt6 and Dt1, Dt3, Dt5, Dt7 of the first 8 QW data from memory rank 461 and 465 simultaneously at time tb. In other words, the accessing module 26 could utilize the double-channel transmission bandwidth to access data when the data is stored in the symmetrically arranged memory ranks.
When accessing the second 8 QW data, the determining module 24 determines that the second 8 QW data is stored in the memory rank 462. Then the accessing module 26 utilizes a control instruction RDY as the control signal S2 of channel 32A to sequentially transfer data blocks Dy0-Dy7 of the second 8 QW data stored in the memory rank 462 at time tb2. In other words, the accessing module 26 could utilize the single-channel transmission bandwidth to access data when the data is stored in the asymmetrically arranged memory rank.
As mentioned above, although the memory 30 is asymmetric, the present invention could still utilize the advantages of the double-channel data transmission for the symmetrically arranged parts of memory ranks of the memory 30. With compared to the prior art, if the memory is asymmetric, one channel is utilized to access data. Taking the example of FIG. 4 and FIG. 5, because the memory 30 is asymmetric, the prior art cannot define the symmetric memory ranks as the same virtual rank. Therefore, the prior art can only utilize one channel (time after timing tb2) to transfer the first 8 QW data (Dt0-Dt7) from the memory rank 461. In other words, the prior art cannot utilize the characteristic “the memory rank 461 has corresponding memory rank 465 with the same memory capacity on the other channel” as well as the present invention.
The third embodiment of the present invention is shown in FIG. 6. Assuming the memory 30 has 4 memory modules 46A, 46B, 48A, and 48B, where the memory modules 46A and 48B are both double-sided and the memory modules 48A and 46B are both single-sided. The memory module 46A has memory ranks 460 and 461, and the memory module 48B has memory ranks 466 and 467. The memory module 48A has one memory rank 462 and the memory module 46B has one memory rank 465. Assuming that the memory rank 460 on channel 32A has the same memory capacity with the memory rank 466 on channel 32B, that is, the memory ranks 460 and 466 are symmetrically arranged. Similarly, the memory rank 461 on channel 32A has the same memory capacity with the memory rank 467 on channel 32B, that is, the memory ranks 461 and 467 are symmetrically arranged. In the present invention, addressing module 28 could define the memory ranks 460 and 466 as a first virtual rank, and define the memory ranks 461 and 467 as a second virtual rank. In addition, data stored in the first virtual rank or stored in the second virtual rank could be accessed through double-channel (similar to the operation of accessing the symmetric memory ranks 461 and 465 shown in FIG. 4). Furthermore, if the memory ranks 462 on channel 32A has different memory capacity with and the memory rank 465 on channel 32B, thus the memory ranks 462 and 465 are asymmetric, thus, data stored in the memory rank 462 or stored in the memory rank 465 could only be accessed through single channel (similar to the operation of accessing the asymmetric memory rank 462 shown in FIG. 4).
The fourth embodiment of the present invention is shown in FIG. 7. Assuming the memory 30 has memory modules 46A and memory module 48B, where memory module 46A is double-side and memory module 48B is single-side. The memory module 46A has two memory ranks 460 and 461 and the memory module 48B has one memory rank 467. Assume that the memory ranks 460, 461, and 467 have the same memory capacity. Because one of the two memory ranks 460 and 461 of the memory module 46B on channel 32A does not have corresponding memory rank on channel 32B, therefore, the memory 30 is asymmetric. Under this condition, addressing module 28 of the present invention defines the memory rank 460 and memory 467 as a virtual rank (or defines memory rank 461 and memory rank 467 as a virtual rank). If a data is accessed from a certain address, the determining module 24 determines the memory space that the address corresponds to. If the address belongs to a memory space provided by the memory rank 461 (or memory rank 460), the accessing module 26 only utilizes channel 32A to access data, so that the single-channel transmission bandwidth is utilized. If the address belongs to a memory space provided by the virtual rank of the symmetric memory ranks 460 and 467 (or memory ranks 461 and 467), the accessing module 26 simultaneously utilizes two channels 32A and 32B to access data, so that the double-channel transmission bandwidth is utilized.
FIG. 8 is a flow chart of accessing an asymmetrically arranged memory according to the present invention. If a first memory rank on first channel has same memory capacity with a second memory rank on second channel, thus defining the first memory rank and the second memory rank as a virtual rank. When data stored in the virtual rank is accessed, the double-channel could be utilized. Conversely, if a third memory rank on first channel has no same memory capacity with a fourth memory rank on second channel, thus when data stored in the third memory rank or in the fourth memory rank is accessed, only single-channel is utilized.
To summarize, in the prior art, if the memory is asymmetrically arranged, the data could be accessed through only one channel with respect to the data stored. In contrast to the prior art, even if the memory asymmetrically arranged, the present invention could define symmetrically arranged parts of the memory ranks of the memory as a virtual ranks. If data is stored in symmetrically arranged memory ranks of the memory, channels corresponding to the symmetrically arranged memory ranks could be simultaneously utilized to transfer data. If data is stored in an asymmetrically arranged memory rank of the memory, the channel corresponding to the asymmetrically arranged memory rank could only be utilized to transfer data. Therefore, the present invention could still take the advantages of the double-channel, even if the memory is not asymmetrical arranged. As a result, in the present invention, the accessing efficiency of the memory could be increased. Furthermore, in the embodiments of the invention, functions of the determining module 24 and the addressing module 28 could be implemented by software (for example, the chipset is used for executing software program codes) or hardware). Therefore, this changes the way of accessing the memory according to whether the address is mapped to the symmetric memory ranks during address decoding, page decoding, and command scheduling.
Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.

Claims (20)

1. A memory accessing method for accessing a memory having a first memory rank connected to a first channel and a second memory rank connected to a second channel, the memory accessing method comprising:
determining the memory capacity of the first memory rank and the memory capacity of the second memory rank;
defining the first memory rank and the second memory rank as a virtual memory rank while the memory capacities of the first memory rank and the second memory rank are the same;
accessing data of the virtual memory rank via both the first channel and the second channel simultaneously, wherein the memory is an asymmetrically arranged memory and at least have a third memory rank connected to the second channel; and
accessing data of the third memory rank via the second channel.
2. The memory accessing method of claim 1, wherein for a write request, the method further comprises dividing data into a plurality of data blocks and writing the data blocks into the virtual memory rank via the first channel and the second channel.
3. The memory accessing method of claim 1, wherein for a read request, the method further comprises reading a plurality of data blocks from the virtual memory rank via the first channel and the second channel; and
interleaving the data blocks to regenerate the data.
4. The memory accessing method of claim 1 wherein the first channel and the second channel have a predetermined transmission bandwidth.
5. The memory accessing method of claim 1, while the memory capacities of the first memory rank and the second memory rank are different, accessing data of either the first memory rank or the second memory rank.
6. The memory accessing method of claim 1 wherein the first memory rank and the second memory rank are asymmetrically plugged in slots.
7. The memory accessing method of claim 1, wherein the memory capacity of the third memory rank is different with that of the first memory rank.
8. A memory accessing apparatus comprising:
a first memory rank;
a second memory rank;
a third memory rank; and
a controller coupled to the first memory rank via a first channel and the second and third memory ranks via a second channel;
wherein the controller defines the first memory rank and the second memory rank as a virtual memory rank while the memory capacities of the first memory rank and the second rank are the same, and accesses data of the virtual memory rank via both the first channel and the second channel simultaneously;
wherein the memory is an asymmetrically arranged memory;
wherein the controller accesses data of the third memory rank via the second channel.
9. The memory accessing apparatus of claim 8 wherein while the memory capacities of the first memory rank and the second memory rank are different, the controller accesses data of either the first memory rank or the second memory rank.
10. The memory accessing apparatus of claim 8, wherein for a write request, the controller divides data into a plurality of data blocks and writes the data blocks into the virtual memory rank via the first channel and the second channel.
11. The memory1 accessing apparatus of claim 8, wherein for a read request, the controller reads a plurality of data blocks from the virtual memory rank via the first channel and the second channel, and interleaves the data blocks as data.
12. The channel memory accessing apparatus of claim 8, wherein the first channel and the second channel have a predetermined transmission bandwidth.
13. The memory accessing apparatus of claim 8, wherein the controller comprises:
an accessing module for accessing data from a memory;
an addressing module coupled to the accessing module for addressing the memory; and
a determining module coupled to the accessing module and the addressing module for determining whether the address of data belongs to the virtual memory rank.
14. The memory accessing apparatus of claim 8 wherein the first memory rank and the second memory rank are asymmetrically plugged in slots.
15. The memory accessing apparatus of claim 8, wherein the memory capacity of the third memory rank is different with that of the first memory rank.
16. A chipset for accessing a memory, wherein the memory comprises a plurality of memory ranks, the chipset comprising:
an accessing module, coupled to the memory via a plurality of channels, the accessing modules accesses data of one of the memory ranks via one of the connected channels;
an addressing module, coupled to the accessing module, wherein some of the memory ranks are plugged in different channels, where the addressing module defines one or more virtual memory ranks; wherein a virtual memory rank comprises a pair of memory ranks of the plurality of memory ranks, each connected to different channels and have the same memory capacities; and
a determining module coupled to the accessing module and the addressing module for determining whether the address of data belongs to the virtual memory rank;
wherein the accessing module accesses data of the virtual memory rank via all the connected channels simultaneously, and accesses data of one of the memory ranks via one of the connected channels while the address of data does not belong to a virtual memory rank of the one or more virtual memory ranks;
wherein the memory is an asymmetrically arranged memory.
17. The chipset of claim 16, wherein for a write request, the data is divided into a plurality of data blocks and the data blocks are written into the virtual memory ranks via the connected channels.
18. The chipset of claim 16 wherein for a read request, a plurality of data blocks are read from the virtual memory rank via the connected channels, and the data blocks are interleaved as data.
19. The chipset of claim 16, wherein each channel has a predetermined transmission bandwidth.
20. The chipset of claim 16 wherein the memory ranks of the virtual memory ranks are asymmetrically plugged in slots.
US10/906,748 2004-06-24 2005-03-04 Method and related apparatus for accessing memory Active US7779215B2 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
TW093118387A TWI299497B (en) 2004-06-24 2004-06-24 Method and related apparatus for accessing memory apparatus
TW93118387A 2004-06-24
TW093118387 2004-06-24

Publications (2)

Publication Number Publication Date
US20050289317A1 US20050289317A1 (en) 2005-12-29
US7779215B2 true US7779215B2 (en) 2010-08-17

Family

ID=35507449

Family Applications (1)

Application Number Title Priority Date Filing Date
US10/906,748 Active US7779215B2 (en) 2004-06-24 2005-03-04 Method and related apparatus for accessing memory

Country Status (2)

Country Link
US (1) US7779215B2 (en)
TW (1) TWI299497B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2012145432A3 (en) * 2011-04-19 2013-01-17 Micron Technology, Inc. Channel depth adjustment in memory systems
US8892844B2 (en) 2011-03-07 2014-11-18 Micron Technology, Inc. Methods of accessing memory cells, methods of distributing memory requests, systems, and memory controllers
US9697884B2 (en) 2015-10-08 2017-07-04 Rambus Inc. Variable width memory module supporting enhanced error detection and correction

Families Citing this family (56)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10013371B2 (en) 2005-06-24 2018-07-03 Google Llc Configurable memory circuit system and method
US8077535B2 (en) 2006-07-31 2011-12-13 Google Inc. Memory refresh apparatus and method
US20080082763A1 (en) 2006-10-02 2008-04-03 Metaram, Inc. Apparatus and method for power management of memory circuits by a system or component thereof
US8386722B1 (en) 2008-06-23 2013-02-26 Google Inc. Stacked DIMM memory interface
US20080028136A1 (en) 2006-07-31 2008-01-31 Schakel Keith R Method and apparatus for refresh management of memory modules
US8327104B2 (en) 2006-07-31 2012-12-04 Google Inc. Adjusting the timing of signals associated with a memory system
US9171585B2 (en) 2005-06-24 2015-10-27 Google Inc. Configurable memory circuit system and method
US8359187B2 (en) 2005-06-24 2013-01-22 Google Inc. Simulating a different number of memory circuit devices
US7386656B2 (en) 2006-07-31 2008-06-10 Metaram, Inc. Interface circuit system and method for performing power management operations in conjunction with only a portion of a memory circuit
US8089795B2 (en) 2006-02-09 2012-01-03 Google Inc. Memory module with memory stack and interface with enhanced capabilities
US8438328B2 (en) 2008-02-21 2013-05-07 Google Inc. Emulation of abstracted DIMMs using abstracted DRAMs
US8081474B1 (en) 2007-12-18 2011-12-20 Google Inc. Embossed heat spreader
US8090897B2 (en) 2006-07-31 2012-01-03 Google Inc. System and method for simulating an aspect of a memory circuit
US8055833B2 (en) 2006-10-05 2011-11-08 Google Inc. System and method for increasing capacity, performance, and flexibility of flash storage
US8619452B2 (en) 2005-09-02 2013-12-31 Google Inc. Methods and apparatus of stacking DRAMs
US8796830B1 (en) 2006-09-01 2014-08-05 Google Inc. Stackable low-profile lead frame package
US8335894B1 (en) 2008-07-25 2012-12-18 Google Inc. Configurable memory system with interface circuit
US8397013B1 (en) 2006-10-05 2013-03-12 Google Inc. Hybrid memory module
US8111566B1 (en) 2007-11-16 2012-02-07 Google, Inc. Optimal channel design for memory devices for providing a high-speed memory interface
US8130560B1 (en) 2006-11-13 2012-03-06 Google Inc. Multi-rank partial width memory modules
US8244971B2 (en) 2006-07-31 2012-08-14 Google Inc. Memory circuit system and method
US7392338B2 (en) 2006-07-31 2008-06-24 Metaram, Inc. Interface circuit system and method for autonomously performing power management operations in conjunction with a plurality of memory circuits
US8041881B2 (en) 2006-07-31 2011-10-18 Google Inc. Memory device with emulated characteristics
US8060774B2 (en) 2005-06-24 2011-11-15 Google Inc. Memory systems and memory modules
US9507739B2 (en) 2005-06-24 2016-11-29 Google Inc. Configurable memory circuit system and method
US9542352B2 (en) 2006-02-09 2017-01-10 Google Inc. System and method for reducing command scheduling constraints of memory circuits
DE102005046134B4 (en) * 2005-09-27 2014-05-15 Qimonda Ag Method for information forwarding with electronic components and correspondingly designed module
DK2458505T3 (en) * 2006-02-09 2014-12-01 Google Inc Memory Circulatory System and method
US9632929B2 (en) 2006-02-09 2017-04-25 Google Inc. Translating an address associated with a command communicated between a system and memory circuits
US7724589B2 (en) 2006-07-31 2010-05-25 Google Inc. System and method for delaying a signal communicated from a system to at least one of a plurality of memory circuits
US8074022B2 (en) * 2006-09-28 2011-12-06 Virident Systems, Inc. Programmable heterogeneous memory controllers for main memory with different memory modules
US8051253B2 (en) * 2006-09-28 2011-11-01 Virident Systems, Inc. Systems and apparatus with programmable memory control for heterogeneous main memory
US9984012B2 (en) 2006-09-28 2018-05-29 Virident Systems, Llc Read writeable randomly accessible non-volatile memory modules
US8949555B1 (en) 2007-08-30 2015-02-03 Virident Systems, Inc. Methods for sustained read and write performance with non-volatile memory
US20080082750A1 (en) * 2006-09-28 2008-04-03 Okin Kenneth A Methods of communicating to, memory modules in a memory channel
US7761624B2 (en) * 2006-09-28 2010-07-20 Virident Systems, Inc. Systems and apparatus for main memory with non-volatile type memory modules, and related technologies
US7761626B2 (en) * 2006-09-28 2010-07-20 Virident Systems, Inc. Methods for main memory in a system with a memory controller configured to control access to non-volatile memory, and related technologies
US7761623B2 (en) * 2006-09-28 2010-07-20 Virident Systems, Inc. Main memory in a system with a memory controller configured to control access to non-volatile memory, and related technologies
US7761625B2 (en) * 2006-09-28 2010-07-20 Virident Systems, Inc. Methods for main memory with non-volatile type memory modules, and related technologies
US8189328B2 (en) * 2006-10-23 2012-05-29 Virident Systems, Inc. Methods and apparatus of dual inline memory modules for flash memory
WO2008055270A2 (en) * 2006-11-04 2008-05-08 Virident Systems, Inc. Writing to asymmetric memory
US20080123305A1 (en) * 2006-11-28 2008-05-29 Smart Modular Technologies, Inc. Multi-channel memory modules for computing devices
US8209479B2 (en) 2007-07-18 2012-06-26 Google Inc. Memory circuit system and method
US8006032B2 (en) * 2007-08-22 2011-08-23 Globalfoundries Inc. Optimal solution to control data channels
US9921896B2 (en) 2007-08-30 2018-03-20 Virident Systems, Llc Shutdowns and data recovery to avoid read errors weak pages in a non-volatile memory system
US8080874B1 (en) 2007-09-14 2011-12-20 Google Inc. Providing additional space between an integrated circuit and a circuit board for positioning a component therebetween
US8621132B2 (en) * 2008-01-08 2013-12-31 Cisco Technology, Inc. System and methods for memory expansion
US8825965B2 (en) 2008-01-08 2014-09-02 Cisco Technology, Inc. System and methods for memory expansion
US8521967B1 (en) 2008-06-24 2013-08-27 Virident Systems, Inc. Network computing systems having shared memory clouds with addresses of disk-read-only memories mapped into processor address spaces
US9513695B2 (en) 2008-06-24 2016-12-06 Virident Systems, Inc. Methods of managing power in network computer systems
EP2441007A1 (en) 2009-06-09 2012-04-18 Google, Inc. Programming of dimm termination resistance values
US9773531B2 (en) 2012-06-08 2017-09-26 Hewlett Packard Enterprise Development Lp Accessing memory
US9195595B2 (en) * 2013-03-21 2015-11-24 Htc Corporation Portable apparatus using multi-core storage mechanism and data access method therefor
US9465735B2 (en) * 2013-10-03 2016-10-11 Qualcomm Incorporated System and method for uniform interleaving of data across a multiple-channel memory architecture with asymmetric storage capacity
TWI692774B (en) * 2019-05-15 2020-05-01 瑞昱半導體股份有限公司 Memory system and method for operating a memory system
CN111984183A (en) * 2019-05-22 2020-11-24 瑞昱半导体股份有限公司 Memory system and operation method of memory system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5619471A (en) * 1995-06-06 1997-04-08 Apple Computer, Inc. Memory controller for both interleaved and non-interleaved memory
US6065092A (en) * 1994-11-30 2000-05-16 Hitachi Micro Systems, Inc. Independent and cooperative multichannel memory architecture for use with master device
US20030088742A1 (en) * 1999-11-10 2003-05-08 Lee Jeffery H. Parallel access virtual channel memory system
US6877076B1 (en) * 2000-09-20 2005-04-05 Broadcom Corporation Memory controller with programmable configuration
US20050114587A1 (en) * 2003-11-22 2005-05-26 Super Talent Electronics Inc. ExpressCard with On-Card Flash Memory with Shared Flash-Control Bus but Separate Ready Lines

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6065092A (en) * 1994-11-30 2000-05-16 Hitachi Micro Systems, Inc. Independent and cooperative multichannel memory architecture for use with master device
US5619471A (en) * 1995-06-06 1997-04-08 Apple Computer, Inc. Memory controller for both interleaved and non-interleaved memory
US20030088742A1 (en) * 1999-11-10 2003-05-08 Lee Jeffery H. Parallel access virtual channel memory system
US6877076B1 (en) * 2000-09-20 2005-04-05 Broadcom Corporation Memory controller with programmable configuration
US20050114587A1 (en) * 2003-11-22 2005-05-26 Super Talent Electronics Inc. ExpressCard with On-Card Flash Memory with Shared Flash-Control Bus but Separate Ready Lines

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
Besedin, Dmitri "RAM FAQ 1.0" May 14, 2006 retrieved from the Internet Dec. 14, 2009 from http://ixbtlabs.com/articles2/mainboard/ram-faq-2006.html. *
Cosair, "Currents:Memory News from Corsair" Aug. 2004, Corsair pp. 1-2. *
Publisher: HP, Title: An overview of system memory technologies, technology brief, 8th edition, Apr. 1, 2009.
Publisher: Kingston, Title: Memory Ranks and Intel E7320/E7520 Chipset-based Servers, Jul. 24, 2008.
Publisher: Kingston, Title: Ultimate Memory Guide-bank, Jul. 24, 2008, printed on Apr. 7, 2010, from http://www.kingston.com/tools/umg/umg02.asp.
Publisher: Kingston, Title: Ultimate Memory Guide-interleaving, Jul. 24, 2008, printed on Apr. 7, 2010, from http://www.kingston.com/tools/umg/umg03.asp.

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8892844B2 (en) 2011-03-07 2014-11-18 Micron Technology, Inc. Methods of accessing memory cells, methods of distributing memory requests, systems, and memory controllers
US9933972B2 (en) 2011-03-07 2018-04-03 Micron Technology, Inc. Methods of accessing memory cells, methods of distributing memory requests, systems, and memory controllers
US10162557B2 (en) 2011-03-07 2018-12-25 Micron Technology, Inc. Methods of accessing memory cells, methods of distributing memory requests, systems, and memory controllers
WO2012145432A3 (en) * 2011-04-19 2013-01-17 Micron Technology, Inc. Channel depth adjustment in memory systems
US11580039B2 (en) 2011-04-19 2023-02-14 Micron Technology, Inc. Channel depth adjustment in memory systems
US10838886B2 (en) 2011-04-19 2020-11-17 Micron Technology, Inc. Channel depth adjustment in memory systems
US10339999B2 (en) 2015-10-08 2019-07-02 Rambus Inc. Variable width memory module supporting enhanced error detection and correction
US10650881B2 (en) 2015-10-08 2020-05-12 Rambus Inc. Variable width memory module supporting enhanced error detection and correction
US10014047B2 (en) 2015-10-08 2018-07-03 Rambus Inc. Memory module supporting time-division memory access
US10878888B2 (en) 2015-10-08 2020-12-29 Rambus Inc. Variable width memory module supporting enhanced error detection and correction
US11164622B2 (en) 2015-10-08 2021-11-02 Rambus Inc. Variable width memory module supporting enhanced error detection and correction
US9697884B2 (en) 2015-10-08 2017-07-04 Rambus Inc. Variable width memory module supporting enhanced error detection and correction
US11705187B2 (en) 2015-10-08 2023-07-18 Rambus Inc. Variable width memory module supporting enhanced error detection and correction
US11967364B2 (en) 2015-10-08 2024-04-23 Rambus Inc. Variable width memory module supporting enhanced error detection and correction

Also Published As

Publication number Publication date
TWI299497B (en) 2008-08-01
TW200601350A (en) 2006-01-01
US20050289317A1 (en) 2005-12-29

Similar Documents

Publication Publication Date Title
US7779215B2 (en) Method and related apparatus for accessing memory
EP1488323B1 (en) Memory system with burst length shorter than prefetch length
US7707366B2 (en) Memory control device
US7451273B2 (en) System, method and storage medium for providing data caching and data compression in a memory subsystem
US7117421B1 (en) Transparent error correction code memory system and method
US5822251A (en) Expandable flash-memory mass-storage using shared buddy lines and intermediate flash-bus between device-specific buffers and flash-intelligent DMA controllers
KR100973629B1 (en) Memory device, method and system having multiple address, data and command buses
CN101040274A (en) Command controlling different operations in different chips
KR100847968B1 (en) Dual-port semiconductor memories
KR20010093642A (en) Improved memory system apparatus and method
KR100679370B1 (en) Memory device having different burst order addressing for read and write operations
CN101836194B (en) An optimal solution to control data channels
CN114207592A (en) Method for error detection and correction and corresponding system and device
EP0775347B1 (en) Integrated primary bus and secondary bus controller with reduced pin count
US9176906B2 (en) Memory controller and memory system including the same
US20100262763A1 (en) Data access method employed in multi-channel flash memory system and data access apparatus thereof
EP0831402A1 (en) Dynamically configuring timing to match memory bus loading conditions
US8099530B2 (en) Data processing apparatus
US20060294327A1 (en) Method, apparatus and system for optimizing interleaving between requests from the same stream
CN115543861A (en) Method for managing address mapping and near memory handling dual inline memory module
US7586779B2 (en) Controller apparatus for utilizing downgrade memory and method for operating the same
US20040034748A1 (en) Memory device containing arbiter performing arbitration for bus access right
KR102583301B1 (en) Dram device and rank mapping method thereof and computer program for the same
KR102545175B1 (en) Operating method of memeory device including address table and memory controller thereof
US8635418B2 (en) Memory system and method for passing configuration commands

Legal Events

Date Code Title Description
AS Assignment

Owner name: VIA TECHNOLOGIES INC., TAIWAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:LIOU, MING-SHI;HSIEH, BOWEI;LAI, JIIN;REEL/FRAME:015729/0912

Effective date: 20050301

STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552)

Year of fee payment: 8

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 12TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1553); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 12